Data-driven Occupancy Profiles for Apartment-style Student Housing

Abstract

To support the need for a better understanding of occupant behaviors in buildings, we present occupancy profiles for apartment-style student housing from high-resolution monitored occupancy data. We illustrate differences between data-driven occupancy patterns and the widely-used reference occupancy schedule in Building America’s House Simulation Protocol. We evaluate the sensitivity of predicted savings associated with the installation of occupancy-based heating and cooling controls to differences in occupancy schedules. Results demonstrate a wide variety of occupancy patterns among students. Weekday occupancy schedules are influenced by students’ class schedules with the average probability that students are home mid-day twice that defined by the reference occupancy schedule. However, the differences between the reference and actual occupancy profiles have produced a small change (~3%) in the savings predicted for occupancy-based heating and cooling controls in a cold climate (4.3–7.7%). The potential savings was not sensitive to the variability among student occupancy schedules due to the high probability of occupancy for most hours of the day. Nonetheless, the variety of occupancy patterns among students reveals diversity in students’ behaviors. Better understanding of occupant behavior in buildings provides crucial information for demand and distribution management of electrical grids, behavioral studies, and occupant-centric building design.